Warm dark matter signatures on the 21cm power spectrum: Intensity mapping forecasts for SKA

TL;DR

This study explores how warm dark matter influences 21cm intensity mapping signals at high redshifts, revealing that WDM models can be distinguished from cold dark matter using SKA observations due to their unique power spectrum signatures.

Contribution

The paper provides hydrodynamic simulations showing that WDM models produce increased 21cm power spectra compared to CDM, and assesses SKA's ability to constrain WDM particle masses.

Findings

WDM models show increased 21cm power despite suppressed matter power spectra.

SKA can rule out 4 keV WDM with 5000 hours at z > 3.

3 keV WDM can be constrained at >2 sigma with 1000 hours at z > 5.

Abstract

We investigate the impact that warm dark matter (WDM) has in terms of 21cm intensity mapping in the post-reionization Universe at z = 3 - 5. We perform hydrodynamic simulations for 5 different models: cold dark matter and WDM with 1,2,3,4 keV (thermal relic) mass and assign the neutral hydrogen a-posteriori using two different methods that both reproduce observations in terms of column density distribution function of neutral hydrogen systems. Contrary to naive expectations, the suppression of power present in the linear and non-linear matter power spectra, results in an increase of power in terms of neutral hydrogen and 21cm power spectra. This is due to the fact that there is a lack of small mass halos in WDM models with respect to cold dark matter: in order to distribute a total amount of neutral hydrogen within the two cosmological models, a larger quantity has to be placed in the most massive halos, that are more biased compared to the cold dark matter cosmology. We quantify this effect and address significance for the telescope SKA1-LOW, including a realistic noise modeling. The results indicate that we will be able to rule out a 4 keV WDM model with 5000 hours of observations at z > 3, with a statistical significance of > 3 sigma, while a smaller mass of 3 keV, comparable to present day constraints, can be ruled out at more than 2 sigma confidence level with 1000 hours of observations at z > 5.